Lars Roemheld
@LarsRoemheld
Followers
427
Following
430
Media
9
Statuses
63
Making machines learn, preferably about causality. Leading data science in Microsoft’s Office of the Chief Economist.
San Francisco, CA
Joined October 2011
I wrote up a short (English) summary of my talk on the German #CoronaWarnApp at #rC3. https://t.co/z03JGsoAft Original talk (in German):
Was haben Flamewars mit dem Silicon Valley und dem Gesundheitswesen zu tun? Beim #rC3 darf ich wieder Gast bei @ChaosWildWest sein und spreche direkt zum Auftakt am 27.12. um 13h. #cwa
7
4
9
How do humans interact with pricing algorithms within firms❓ We explore this question using two large-scale field experiments in collaboration with @ZalandoTech in a new WP. This work is a collaboration with Tobias Huelden, @VJascisens and @LarsRoemheld. 🧵Thread below.
3
20
70
"I find that 'applied science' has different expectations than research science. So I wrote up what, in my experience, helps produce great applied science work." by Lars Roemheld
1
4
18
A big part of leadership is making expectations explicit. Over the last year, I've formalized what I believe to be great applied (data) science work. Whether as a "definition of done", inspiration for personal development, or ideas to shape team culture. https://t.co/tWJwOlk697
9
2
9
I don't know if AI will kill us all but this already seems like a crime against humanity
1
1
1
Exciting: a feature-rich, fast GLM in python!
I'm excited about the ability to set penalties for individual elastic net parameters. And to have variable matrix-valued Tikhonov matrix penalties. This lets you do cool things like imposing smoothness between adjacent zip-code fixed effects.
2
0
2
"As I ate the still-living digital body of my friend in the company of a stranger in the shape of a wild dog, I thought, 'Oh. This is **special**.'" Deeply beautiful essay about finding home and meaning on the internet by @everestpipkin
https://t.co/TnvvsRJwGx
pioneerworks.org
Beauty and solace in the abandoned worlds of Roblox.
0
0
0
Meldewochen 2020/2021 im Vergleich: Die 7-Tage-Inzidenz zeigt starken Anstieg bei den 15-34-Jährigen.😮 Angesicht der #DeltaVariante ist es wichtig, die #CoronaWarnApp (empfohlen >16) jetzt zu nutzen. Ihr könnt im Infektionsfall andere direkt warnen. ➡️ https://t.co/iW9OLFcLIn
16
74
210
VQGAN + CLIP "matte painting of someone reading papers and burning the midnight oil | trending on artstation"
2
17
165
... it neatly summarizes what's happening: using 2 RandomForests, confounding noise is reduced. This comes at a cost, as we explain away and therefore lose variation in our cause of interest.
0
0
0
Using @VC31415 's Double Machine Learning for causal inference in industry: I wrote up a deep dive article about a real-world example. I like this plot, because... 👇 https://t.co/NG0JsPVlHH
5
1
6
Auch für #DeltaVariante: Es gibt inzwischen gute Hinweise, dass die #CoronaWarnApp funktioniert! Mehr freiwillige Warnungen = langsamere Verbreitung. Mehr-Nutzer bringen wegen Netzwerkeffekten überproportional Mehr-Nutzen. Installieren hilft.
Digitale Werkzeuge können helfen, die #Pandemie einzudämmen. Die Leopoldina hat heute ein Diskussionspapier zum Einsatz automatisierter #Kontaktnachverfolgung, #CoronaApps und #Wearables veröffentlicht: https://t.co/vXMjRCNL0F
#Corona #Coronavirus
2
2
1
Many important points. #1 issue is missing though: reliable antigen/antibody tests are cheap. Why put somebody in expensive scanners for the same diagnostic quality in the first place?
Can you detect COVID-19 using Machine Learning? 🤔 You have an X-ray or CT scan and the task is to detect if the patient has COVID-19 or not. Sounds doable, right? None of the 415 ML papers published on the subject in 2020 was usable. Not a single one! Let's see why 👇
2
0
3
What a time to be alive.
Hallo @einaugenschmaus, vielen Dank für Ihre Anfrage. David Hasselhoff hat keine Gage erhalten und hat sich aus Überzeugung mit seiner Botschaft beteiligt. Wir bitten Sie, sich auf geprüfte Informationen zu stützen. Bleiben Sie gesund.
1
0
1
Don't touch the collider, and be clever about observational, interventional, and randomized data: very much enjoyed teaching a practitioner's perspective on causal inference. Thanks for having me!
What an awesome opportunity our @helmholtz_ai provided me today: a full day workshop on causal inference by 2 superb teachers & formidable experts @k__niki @LarsRoemheld! I am super grateful 🙏 and overwhelmed at the same time. ☕️ Mind the colliders! 💥⚠️
1
0
1
24.300 bestätigte Neuinfektionen. Viel zu viele! Und 4.449 warnende Personen über die Corona-Warn-App, die offensichtlich besser funktioniert als ihr Talkshow-Image verspricht.
6
57
264
The year we learned that people don’t think in MapReduce
2015: NoSQL! 2021: No, SQL!
2
0
2
Um 11:00 Uhr auf @joinclubhouse diskutiere ich mit @christiansauer @gadgetman82 Harald Eichsteller @kianamardi & @LarsRoemheld über die Bedeutung der Corona-Warn-App in der Pandemiebekämpfung. Seid dabei! https://t.co/lbbRRpBDtj
3
4
11
Die Geschichte der #CoronaWarnApp, erzählt beim #rc3 , kann man Dank der super Arbeit von @ChaosWildWest schon jetzt im Stream (nach)schauen. https://t.co/hbZ3EdePuh
Look 'ma, I'm on #Tagesschau! 1 second of fame for my talk on #CoronaWarnApp . Download the app for your friends and family, and remember that for warnings you need to opt-in twice; on the lab order and in the app. (h/t @naehrdine for being the main character!)
2
2
24